ArticlePDF Available

Cheatgrass die-offs as an opportunity for restoration in the Great Basin, USA: Will local or commercial native plants succeed where exotic invaders fail?


Abstract and Figures

Bromus tectorum (cheatgrass) has widely invaded the Great Basin, U.S.A. The sporadic natural phenomenon of complete stand failure (‘die-off’) of this invader may present opportunities to restore native plants. A recent die-off in Nevada was precision-planted with seeds of the native grasses Poa secunda (Sandberg bluegrass) and Elymus elymoides (bottlebrush squirreltail), of both local and nonlocal origin, to ask: 1) Can native species be restored in recent B. tectorum die-offs? And 2) Do local and nonlocal seeds differ in performance? Additionally, we asked how litter removal and water addition affected responses. Although emergence and growth of native seeds was lower in die-off than control plots early in year one, in year two, seedlings in die-offs had increased vigor and growth, at equal or higher densities, than control plots. Local seeds consistently outperformed nonlocal seeds for P. secunda, whereas for E. elymoides, nonlocal showed an advantage in the first season, but in the second season, there were more local seeds present under die-off and unraked conditions. Seedbed treatments affected performance, but did not notably improve establishment or modify other results. Our results warrant further investigation into die-off restoration as well as recognition of the importance of seed source selection in restoration.
Content may be subject to copyright.
Cheatgrass die-offs as an opportunity for restoration in the Great
Basin, USA: Will local or commercial native plants succeed where
exotic invaders fail?
Owen W. Baughman
, Susan E. Meyer
, Zachary T. Aanderud
, Elizabeth A. Leger
Department of Natural Resources and Environmental Science, University of Nevada Reno, 1664 N. Virginia St., Reno, NV 89557, USA
U.S. Forest Service Rocky Mountain Research Station, Shrub Sciences Laboratory, 735 North 500 East, Provo, UT 84606, USA
College of Life Sciences, Brigham Young University, 4125 LSB, Provo, UT 84602, USA
article info
Article history:
Received 19 March 2015
Received in revised form
29 July 2015
Accepted 7 August 2015
Available online 29 August 2015
Bromus tectorum
Stand failure
Local adaptation
Poa secunda
Elymus elymoides
Bromus tectorum (cheatgrass) has widely invaded the Great Basin, U.S.A. The sporadic natural phe-
nomenon of complete stand failure (die-off) of this invader may present opportunities to restore native
plants. A recent die-off in Nevada was precision-planted with seeds of the native grasses Poa secunda
(Sandberg bluegrass) and Elymus elymoides (bottlebrush squirreltail), of both local and nonlocal origin, to
ask: 1) Can native species be restored in recent B. tectorum die-offs? And 2) Do local and nonlocal seeds
differ in performance? Additionally, we asked how litter removal and water addition affected responses.
Although emergence and growth of native seeds was lower in die-off than control plots early in year one,
in year two, seedlings in die-offs had increased vigor and growth, at equal or higher densities, than
control plots. Local seeds consistently outperformed nonlocal seeds for P. secunda, whereas for
E. elymoides, nonlocal showed an advantage in the rst season, but in the second season, there were more
local seeds present under die-off and unraked conditions. Seedbed treatments affected performance, but
did not notably improve establishment or modify other results. Our results warrant further investigation
into die-off restoration as well as recognition of the importance of seed source selection in restoration.
©2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND
license (
1. Introduction
The introduction and spread of invasive exotic species is an
ongoing global issue with broad and negative impacts on biodi-
versity, ecosystem integrity, and essential ecosystem services on
which human well-being depends (Py
sek and Richardson, 2010).
Within the United States, exotic species invasion is second only to
habitat loss and degradation as a leading cause of biodiversity
reduction (Wilcove et al., 1998). One such species, the annual grass
Bromus tectorum L. (cheatgrass), has come to colonize tens of mil-
lions of acres of cold desert shrublands since introduction more
than a century ago (Mack, 1981; Knapp, 1996). Because of its effects
on ecosystem processes (Brooks et al., 2004;Adair and Burke, 2010;
Beckstead et al., 2010) and strong competitive effects on native
species (Nasri and Doescher 1995; Rafferty and Young 2002).
B. tectorum invasion can dramatically limit land management op-
tions and make restoring more desirable species challenging
(Davies et al., 2011).
Bromus tectorum die-off is a common but poorly understood
phenomenon in which an abundant B. tectorum seed bank fails to
produce a stand of living plants, despite receiving precipitation
that is sufcient for establishment. While some areas experiencing
die-off continue to show stand failure for several years, many areas
recover to considerable B. tectorum densities the following year
(Baughman, 2014). The factors directly responsible for B. tectorum
die-off have yet to be determined, though recent studies have
implicated several fungal pathogens that demonstrated high
pathogenicity in the laboratory under specic conditions (Meyer
et al., 2014; Meyer pers. com.). Such naturally occurring and
complete stand failures of dominant annual species in arid or
semi-arid wildlands are unprecedented in the literature. However,
similar die-off processes do occur in agricultural settings,
including take-all fungus of wheat (Bithell et al., 2011), sudden
*Corresponding author. University of Nevada Reno, 1664 N. Virginia St., MS186,
Reno, NV 89557, USA.
E-mail address: (O.W. Baughman).
Contents lists available at ScienceDirect
Journal of Arid Environments
journal homepage:
0140-1963/©2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (
Journal of Arid Environments 124 (2016) 193e204
death syndrome of soybeans (Hartman et al., 2015), and maize
streak virus of corn (Shepherd et al., 2010). The existence of these
examples of pathogens causing stand failure in agricultural sys-
tems suggests that this mechanism is possible in other high den-
sity, low diversity, annual systems, such as B. tectorum near-
Because these B. tectorum die-offs represent a sudden but
temporary decrease of a highly competitive exotic, they could be an
opportunity for restoration if native species can establish in these
altered conditions. Following die-off, soils show a many-fold in-
crease in soil mineral nitrogen (N) relative to immediately adjacent
soils not exhibiting die-off (Blank et al., 2011; Meyer et al., 2014).
This abundant soil nitrogen and reduced competition with
B. tectorum in die-offs could aid in native establishment, or alter-
nately, the causal agents of the die-off may cause mortality of native
species seeded into die-offs. This approach of investigating win-
dows of restoration opportunity may be of use in other situations
where large disturbances have temporarily removed undesirable
and/or invasive species, such as wildre, herbicide applications,
recent agricultural abandonment or fallowing, or successful
biocontrol efforts.
Restoring native plant communities to B. tectorum-invaded
lands is made challenging not only by the presence of the invader,
but also due to incomplete knowledge currently guiding man-
agement practices in several important areas (Davies et al., 2011).
One such area is the importance of seed source in restoration
(Hardegree et al., 2011). Local adaptation, or higher tness of local
than non-local genotypes when evaluated in local conditions, is
common in many plant communities (Clausen et al., 1947;
Loveless and Hamrick, 1984; Linhart and Grant, 1996; Kawecki
and Ebert, 2004), including the Great Basin (e.g. Meyer et al.,
1995; Rice and Davis, 2009; Rowe and Leger, 2012; Johnson
et al., 2013). Despite this, seed demand in the Great Basin is
typically met with commercially-produced native seed derived
from programs aiming to create widely-adapted, workhorse
genotypes (United States Department of Agriculture, 2013). Un-
fortunately, the portions of the Great Basin most at risk of
degradation are poorly represented by the collection locations of
many of these commercial seed lots (Jensen and Stettler, 2012),
and thus for much of the Great Basin, seeds used in restoration are
likely to lack traits that are adapted to local biotic or abiotic
conditions, including the die-off phenomenon. It is vital that more
research compare the success of local and nonlocal native plants
in the Great Basin, especially in highly invaded and potentially
modied systems, such as those containing B. tectorum near-
To examine whether die-offs are potential restoration oppor-
tunities, this study employed a two-year, in situ seed establish-
ment experiment within a naturally occurring die-off in
northecentral Nevada using two common grass species native to
the Great Basin, Poa secunda J. Presl. (Sandberg bluegrass) and
Elymus elymoides (Raf.) Swezey (bottlebrush squirreltail). We
address the following question: 1) Can native species be success-
fully restored in a recent B. tectorum die-off? Additionally, seeds of
both local and nonlocal (commercial) origin were included in the
experiment to address the question: 2) Do native plants of local
and nonlocal origins differ in performance, and if so, are these
differences consistent in and out of a recent die-off? Further, we
included early-season water addition and litter removal treat-
ments to determine if such ameliorations are necessary for
establishing native plants in die-off areas. We also documented
soil fertility, soil moisture, soil temperature, and the abundance of
B. tectorum in die-off and infested areas to determine if these
factors were associated with any differences in native plant
2. Methods
2.1. Site selection and description
Extensive surveys in Northern Nevada, guided by knowledge of
previous die-off areas as well as by preliminary recommendations
generated through a remote sensing effort aimed at detecting die-
offs (Weisberg pers. com.), revealed only a single die-off site in
spring 2012 in northern Buena Vista Valley, northecentral Nevada
(Fig. 1). The study site is located at 1384 m elevation, and the 30-
year mean annual precipitation is 222 mm, and mean annual
temperature is 9.8
C(PRISM Climate Group, 2014). Formerly
dominated by Artemisia tridentata Nutt. ssp. wyomingensis Beetle &
Young (Wyoming big sagebrush), the site was last affected by
wildre in 1999. It is shallowly sloping and has well-drained,
relatively deep ne sandy loam soils (Natural Resources
Conservation Service Soil Survey Staff, 2014). At the time of the
study, the general area was occupied primarily by B. tectorum, with
low densities of Sisymbrium altissimum L. (tall tumble mustard),
Salsola tragus L. (Russian thistle), Lepidium perfoliatum L. (clasping
pepperweed), E. elymoides and Microsteris gracilis (Hook.) Greene
(slender phlox). B. tectorum die-offs were rst observed and
investigated in the area in 2008 (Baughman and Meyer, 2013). An
area (70 m 50 m) that contained a representative sample of the
recent die-off as well as an adjacent, unaffected area with an intact
B. tectorum stand (control) was selected for the experiment and was
fenced to exclude grazing. The die-off area supported virtually no
B. tectorum growth during the growing season prior to installation
of the experiment (2012), while the adjacent controlarea sup-
ported a stand of B. tectorum. September 2012 seed bank samples
showed no signicant (P >0.05) differences between die-off and
control conditions in B. tectorum seeds per square meter, for viable
(6200 ±1500 SE in control, 10,200 ±4500 SE in die-off) or nonvi-
able seeds (12,300 ±2000 SE in control, 10,300 ±2200 SE in die-
off), indicating that it was in fact a recent, rst year die-off
(Baughman and Meyer, 2013). Site precipitation was estimated to
be 56% of average in the 2012 water year (October 1
2011eSeptember 30 2012) during which the die-off occurred, as
well as 69% and 92% in the 2013 and 2014 water years, respectively,
during which the experiment took place.
2.2. Species selection, seed acquisition and processing
The cool-season, perennial grasses selected for this study were P.
secunda (Sandberg bluegrass) and E. elymoides (bottlebrush squir-
reltail), which were found growing in natural populations within 2
km of the site. P. secunda is a widely distributed bunchgrass that
generally grows and matures earlier in the growing season than
most other native perennial grasses. It is considered a pioneer
species and is relatively tolerant of re and other disturbances. E.
elymoides is also a widely distributed bunchgrass, and matures
slightly later than P. secunda. These species were selected because
they commonly grow in similar habitats and are frequently used in
restoration in the region. Additionally, both species have demon-
strated abilities to compete with or tolerate B. tectorum (Booth et al.,
2003; Goergen et al., 2011; Stevens et al., 2014).
Mature seeds of local collections were hand-harvested
throughout June and early July, 2012 from as many individuals
(>50) and as close to the study site (<6 km) as possible. Seeds were
stored at room temperature in paper envelopes for 6 months.
Commercially produced, nonlocal seeds were obtained from L&H
Seeds, a private seed grower in eastern Washington. Mt. Home
germplasm P. secunda (Lambert et al., 2011) was chosen because its
origin in southern Idaho was geographically closer and more
environmentally similar to the study site than any other commonly
O.W. Baughman et al. / Journal of Arid Environments 124 (2016) 193e204194
available germplasm at the time. Toe Jam Creek germplasm
E. elymoides (Jones et al., 2004), originating from northern Nevada,
was selected for similar reasons. The origins of the Mt. Home and
Toe Jam Creek germplasms experience an average of 273 mm and
322 mm annually, respectively, compared to 222 mm at the
planting site where local germplasms were collected (Fig. 1,PRISM
Climate Group, 2014), but conditions at the commercial growing
site are unknown. Both local and nonlocal seed lots were hand-
sorted on a light table, and only seeds with well-developed endo-
sperm were selected. This sorting was required as seed set at the
local site was low, presumably due to water limitations during the
2011e2012 growing season. The viability of each hand-sorted lot
was assessed through a 30-day, room temperature germination
trial of 25 seeds in each of 8 petri dishes per seed type.
2.3. Field design
Seeds were directly planted into both die-off and adjacent non
die-off (henceforth, control) conditions. Four seedbed treatments
were implemented, varying the micro-environments under which
native establishment could occur. These treatments were designed
to determine if native seed establishment was dependent on
amelioration of any site factors, such as excessive litter or dry
conditions. Treatments were: seeding into unaltered plot and
seeding with litter removed (raked), each with and without arti-
cial precipitation addition (watered). Litter was removed by
gentle hand-raking, and water was added using a plastic bag
perforated with 5 small holes suspended above the plot on hard-
ware cloth and a wooden frame.
Ten blocks were established in the experiment. Each block was
composed of two arrays, with one in the die-off condition and the
other in the control condition, each beginning a minimum of 5 and
a maximum of 12 m from the die-off edge. Within each array, 16
square 0.1 m
plots representing factorial combinations of the four
treatments, two species, and two seed origins were randomly ar-
ranged in two rows of 8, with 15 cm spacing between plots and
block edges (see Baughman, 2014). Using Titebond 5062 Original
wood glue, seeds were attached by their lemmas, with embryo
down, to 15 cm long bamboo skewers at 5 cm from skewer tip and
inserted so that seeds were 0.6e1.2 cm below the soil surface. This
method allowed for condent differentiation of target seedlings
from resident seedlings, and very high seedling detectability
through time. Each plot contained a four by ve matrix of seeds on
skewers (20 seeds total), all from one of the four seed lots, planted
5e6.5 cm from one another. This design of ten planting blocks
contained a total of 320 plots covering 32 m
planted with 6400
seeds. Soil moisture and temperature measurements were taken in
all treatment combinations in each of three additional blocks,
interspersed between planting blocks. These soil monitoring plots
received the same seedbed manipulations and seeding density as
seeded plots, but were not monitored for seedling activity.
Seeds were planted during the last week of October, 2012.
Seeded plots and soil monitoring plots assigned to the articial
precipitation treatment were watered with ~2.5 cm at one and
three weeks post-planting. After mid-November, moisture to all
plots was dependent upon natural precipitation.
2.4. Data collection
Seeds were carefully examined for emergence in November,
December, and monthly from FebruaryeMay of the rst growing
season (2012e2013) and newly-emerged seedlings were marked
with colored paper clips. All seedlings that had at any point shown
active growth in the rst season were monitored in November,
December, February, April and May of the second growing season
(2013e2014). In both seasons, plants were showing senescence
(dormancy) by the end of May. Data are reported in terms of
actively growing seedlings rather than living seedlings because it
was not possible to differentiate seedling mortality from
All plants exhibiting active growth in early May were measured
in late May (even if senesced) for leaf number and longest leaf
length (hereafter, height), and given a greenness integer rating
(hereafter, late-season vigor) from 0 (0% green) to 3 (75e100%
green). In the second year, the number of plants in each plot pro-
ducing owering structures was recorded. B. tectorum density was
recorded in each plot in early May 2013, and aboveground
B. tectorum biomass was harvested from within plots in August
2013 after most seeds had shattered. Biomass samples were dried
in an oven at 40
C for at least two days and weighed.
Fig. 1. An example of a Bromus tectorum die-off in Pershing County, NV in 2014. Bromus tectorum grew in a near-monoculture on the land in both sides of this image in 2013, but
experienced stand failure on the right during the 2014 growing season. The gray litter layer on the right is typical of B. tectorum die-offs, and is in contrast to the standing, healthy
stand of B. tectorum on the left. The stark edge between these two conditions is not delineated by obvious topographical, edaphic, or climatic boundaries. Inset climate graphs for the
study location/local germplasm origin (40.6882N, 117.9582W) and nonlocal germplasm origins for Mt. Home P secunda (42.3127N, 115.5821W) and Toe Jam Creek E. elymoides
(41.3127N, 116.4738W) are based on 30-year averages (PRISM Climate Group, 2014).
O.W. Baughman et al. / Journal of Arid Environments 124 (2016) 193e204 195
To determine the effects of experimental factors on soil condi-
tions, real-time soil moisture (cm
soil) and temperature
C) were measured with ECH
O-TM sensors and Decagon data
loggers (Decagon Devices, Pullman WA, USA). Sensors were placed
at 2 cm below soil surface in all treatment combinations in each of
three blocks (for a total of 24 sensors). All values were expressed on
a daily time-step to remove diel variability. Moisture and temper-
ature data were separated into three time periods: winter (1st Nov
e28th February), early spring (1st March e30th April), and late
spring (1st May e31st May), with these divisions based on the
timing of spring-thaw (end of February) or stable warm soil tem-
peratures (>15
C beginning of May) in the rst growing season
(Gornish et al., 2014).
Soil samples were taken before the growing season, in October
2012, and after the growing season in June 2013 in 6 cm
diameter 8 cm deep cores. October samples were a composite of
5 randomly-located cores per planting array from ve blocks,
stored in paper bags and frozen for 10 months, while June samples
were a composite of 2 random cores per array from all ten blocks,
and stored at room temperature in paper bags for several weeks.
One composite sample from each array and sample date was
analyzed for pH, organic matter, nitrate, phosphorus, potassium,
calcium, and magnesium (A&L Analytical Laboratories, Memphis,
2.5. Data analysis
Cumulative emergence, seedling number, and number of plants
producing owering structures were calculated as percentages of
the total seed planted on a per-plot basis. These percentages were
then adjusted for the viability of each hand-sorted seed lot
(P. secunda: 70% for nonlocal and 77.5% for local; E. elymoides: 96.3%
for nonlocal, 98.8% for local) to normalize responses as proportions
of the total viable seed planted. Response variables included cu-
mulative emergence (sum of seedlings per plot that emerged at any
point during the experiment), proportion of viable seeds actively
growing at each of eleven sampling dates, mean plot plant height,
leaf number, and late-season vigor in May of both seasons, mean
plot number of owering plants in the second season, and rst
season B. tectorum plot density and biomass. Data were analyzed
separately for each species, using a mixed model ANOVA with
condition (CND: control, die-off), seedbed treatment (TRT: none,
raked), water treatment (WTR: unwatered, watered), and origin
(ORGN: local, nonlocal) as xed factors, with all possible in-
teractions, and BLOCK as a random factor (JMP v10.0, SAS Institute
Inc.). Initially, data were analyzed as a repeated measures (RM)
ANOVA, with TIME (11 sampling dates) included as a xed and fully
crossed factor in the above model for the active growth response,
but all factors had highly signicant interactions with TIME for both
species (analysis not shown). Therefore, the above model was run
separately for each sampling date. Late-season vigor, height, leaf
number, and second season owering were analyzed with
nonparametric Wilcoxon signed rank tests without BLOCK for only
the factors CND and ORGN due to small sample sizes and unequal
variance that precluded analysis of these responses with the
ANOVA model described above. The effects of the treatments on
soil moisture and temperature for the rst season time periods
winter, early spring, and late spring were evaluated with RM
ANOVA with CND, TRT, WTR, and TIME as xed factors and BLOCK
as a random factor in the R Statistical Environment(R Development
Core Team, 2013). Soil data were also analyzed with ANOVA, with a
model containing CND as a xed factor and BLOCK as a random
factor. Response variables were transformed where necessary to
improve residual normality and homogeneity of variance, and
Tukey's HSD tests were used to determine differences for
signicant ANOVA interactions.
To test whether differences in B. tectorum dynamics may have
contributed to differences in seedling performance between die-off
and control sites, we conducted additional analyses that included
plot-level B. tectorum biomass and density individually as cova-
riates in the main model described above. Both B. tectorum mea-
sures were used because they were not strongly correlated
¼0.10). We report both the signicance of the relationship
between B. tectorum dynamics and response variables, as well as
changes in signicance of CND effects and CND interactions,
focusing on rst season emergence and April and May active
seedling number in the rst year as responses for this test. Changes
in signicance of effects were interpreted as follows: if, for
example, there was a signicant CND effect in the main model, but
including B. tectorum density as a covariate eliminated this effect,
this is consistent with the hypothesis that differences in B. tectorum
density between conditions may have been responsible for any
observed differences in plants grown in die-off vs. control condi-
tions. Note, however, that we did not manipulate B. tectorum den-
sities as part of this experiment, and these analyses present only
correlative patterns, and the results could indicate plant responses
to factors other than competition.
Unless otherwise noted, data are presented as plot means ±SE,
and signicance of results is interpreted as P values <0.05.
3. Results
3.1. Soil resources, moisture, and temperature
Both before and after the rst growing season, die-off soils had
signicantly lower pH, greater P, and greater NO
than adjacent
control soils, and after the growing season, die-off soils had
signicantly lower Mg (Table A.1). Soil in the die-off was continu-
ally wetter in winter and early spring, as well as cooler in winter,
early, and late spring (Table A.2,Figure A.1). Litter removal and
water addition affected soil moisture and temperature, with some
differences in these effects in the die-off vs. control (Table A.2,
Appendix 1).
3.2. Bromus tectorum dynamics
Density (plants/0.1 m
)ofB. tectorum in May 2013 was signi-
cantly lower in die-off than control (die-off: 14.4 ±1.7 SE, control:
22.2 ±1.4 SE; F ¼20.6
,P¼0.001), in raked than in unraked
treatment (raked: 12.7 ±1.0, unraked: 24.1 ±1.8 SE; F ¼50.9
P<0.001), and in unwatered than watered treatment (unwatered:
14.1 ±1.1 SE, watered: 22.8 ±1.8 SE; F ¼9.1
,P¼0.01). Though
densities were lower, biomass (g/0.1 m
)ofB. tectorum in August
2013 was signicantly higher in the die-off than the control across
most treatments (die-off: 6.6e9.7 ±0.6e0.9 SE, control:
4.0e5.4 ±0.3 SE; F ¼13.3
,P¼0.005), with the exception of
unwatered plots in the raked treatment, where there were no
between-condition differences (TRTWTRCND interaction;
,P<0.001). Watering generally increased B. tectorum
biomass and raking generally decreased biomass, with some ex-
ceptions in some treatments (data not shown, see Baughman,
2014). Neither the native species planted nor their origin had sig-
nicant effects upon B. tectorum density or biomass.
3.3. Emergence timing and cumulative emergence
Seeds of both species that received additional water emerged
earlier than unwatered seeds, showing more active growth from
November 2012 through Feb 2013 (Tables 1 and 2,Fig. 2), with
some different responses to water based on site condition, seedbed
O.W. Baughman et al. / Journal of Arid Environments 124 (2016) 193e204196
treatments, and origin of seeds; these interactions are discussed
below. Seeds in unwatered treatments remained mostly ungermi-
nated until March 2013; the majority of emergence was completed
by March 2013 for P. secunda and April 2013 for E. elymoides (Fig. 2).
First season cumulative emergence was signicantly higher in
control than die-off for P. secunda (Table 1,Fig. 3a). Cumulative
emergence was signicantly higher in watered E. elymoides control
than in watered die-off, but when unwatered, there was no differ-
ence between conditions (Table 2 CNDWTR interaction, Fig. 3b).
Cumulative emergence was signicantly higher for local than
nonlocal P. secunda, and signicantly higher for nonlocal than local
E. elymoides (Tables 1 and 2,Fig. 3a and b). Water addition was
associated with signicantly higher cumulative emergence of both
species, and litter removal did not signicantly affect cumulative
emergence of either native species (Tables 1 and 2).
3.4. Effects of die-off and treatments on P. secunda active growth
The die-off supported signicantly fewer P. secunda seedlings
than the control in the watered treatment, but not in the unwatered
treatment, from November 2012 to February 2013 (Table 1
CNDWTR interactions, Fig. 2a), and regardless of other factors in
March 2013. The unraked treatment had fewer seedlings in the die-
off than in the control in November 2012 and April 2013
(November: 18% ±3% SE in control, 8% ±2% SE in die-off; April:
61% ±3% SE in control, 43% ±3% SE in die-off), with the raked
treatment showing no condition differences in active seedlings for
these months (Table 1 CNDTRT interactions). Water addition had
Table 1
Full factorial ANOVA results for P. secunda, for rst season (Nov. 2012eMay 2013) cumulative emergence (CE) and rst and second season (Nov. 2013eMay 2014) mean
proportion of viable seedlings showing active growth. Values are F statistics (numerator, denominator degrees of freedom 1, 9 for main factors, 1, 98e99 for interactions), and
bolded effects are signicant (P <0.05), and bold italics indicates P <0.06. Factors include condition (CND, control vs. die-off), seedbed treatment (TRT, raked vs. unraked),
water treatment (WTR, watered vs. unwatered) and seed origin (ORGN, local vs. nonlocal).
CE 2012 2013 2014
Mar Apr May
CND 10.9** 12.3** 7.6* 5.3* 12.0** 9.1* 49.5*** 0.2 0.3 0.0 4.8
TRT 0.1 2.9 0.2 1.2 0.8 0.3 69.7*** 8.5* 16.8** 16.1** 37.6***
WTR 8.7* 79.3*** 96.7*** 129*** 3.0 0.0 0.1 2.5 2.3 4.5 1.1
ORGN 141.5*** 5.6* 8.0* 7.9* 101*** 79.2*** 28.6*** 2.2 9.0** 10.1* 45.8***
CNDTRT 2.2 5.8* 3.3 1.4 3.6 4.5* 0.3 0.0 0.1 0.0 0.3
CNDWTR 1.2 9.3** 17.3*** 18.4*** 0.1 0.3 0.1 1.9 1.2 2.8 2.0
TRTWTR 0.2 4.8* 0.0 0.0 0.2 0.1 0.8 2.2 4.7* 0.2 6.1*
CNDTRTWTR 0.3 1.2 0.1 0.3 1.2 1.4 2.0 0.0 0.1 0.0 0.9
CNDORGN 0.3 0.3 1.1 0.2 0.1 0.1 0.4 0.0 0.9 0.0 0.1
TRTORGN 3.0 0.3 0.8 0.2 1.4 1.4 5.2* 1.3 1.8 1.6 0.5
CONDTRTORGN 0.3 0.3 0.0 0.1 0.3 0.2 2.7 0.5 0.7 0.9 0.4
WTRORGN 1.3 5.1* 7.3** 11.7*** 0.6 0.5 3.0 0.2 0.7 0.9 7.6**
CONDWTRORGN 0.1 0.5 0.0 0.2 0.0 0.0 0.9 0.5 0.1 1.1 0.1
TRTWTRORGN 0.1 0.5 0.0 0.1 0.7 0.7 1.0 0.2 0.1 0.0 0.5
CNDTRTWTRORGN 1.2 0.6 0.1 0.5 0.8 1.2 0.1 0.0 0.4 0.1 0.2
P (0.055), *P (0.05e0.01), **P (0.009e0.001), ***P(<0.001).
ln(y þ0.025) transformation.
Arcsin(y) transformation.
Best BoxeCox Y transformation.
Table 2
Full factorial ANOVA results for E. elymoides, for rst season (Nov. 2012eMay 2013) cumulative emergence (CE) and rst and second season (Nov. 2013eMay 2014) mean
proportion of viable seedlings showing active growth. Values are F statistics (numerator, denominator degrees of freedom 1, 9 for main factors, 1,99 for interactions), and
bolded effects are signicant (P <0.05). Factors include condition (CND, control vs. die-off), seedbed treatment (TRT, raked vs. unraked), water treatment (WTR, watered vs.
unwatered) and seed origin (ORGN, local vs. nonlocal).
CE 2012 2013 2014
Mar Apr May Nov
CND 10.3* 3.1 5.7* 6.5** 19.7** 8.5* 3.2 20.2** 14.7** 17.4** 17.0***
TRT 0.2 23.0** 9.6* 12.6** 1.2 0.2 37.7*** 0.0 0.2 1.6 0.9
WTR 87.4*** 154*** 417*** 648*** 186*** 86.4*** 47.4*** 2.3 2.2 0.8 0.9
ORGN 7.13* 0.0 0.7 1.1 0.0 8.7* 40.1*** 0.6 0.0 0.8 2.1
CNDTRT 1.1 8.0** 0.2 0.1 0.6 1.8 1.2 1.9 0.6 0.5 2.1
CNDWTR 5.8* 19.5*** 47.4*** 41.9*** 0.0 6.7* 2.8 0.7 0.0 0.0 0.1
TRTWTR 2.1 16.3*** 0.0 5.4* 2.3 2.2 12.3*** 1.1 0.1 0.5 0.6
CNDTRTWTR 2.3 11.2** 1.3 0.0 1.5 1.8 0.9 0.6 0.4 1.4 0.2
CNDORGN 0.7 3.2 2.1 2.2 3.4 0.7 0.8 5.2* 8.0** 2.8 4.3*
TRTORGN 0.0 0.0 0.6 1.1 0.0 0.0 0.2 2.6 8.0** 7.3** 4.4*
CONDTRTORGN 2.3 0.7 2.4 0.2 1.2 1.0 0.0 0.1 2.0 4.3* 3.2
WTRORGN 0.3 0.1 1.2 1.8 0.9 0.2 0.0 2.4 0.7 0.3 1.5
CONDWTRORGN 2.3 5.7* 11.6*** 5.2* 0.0 1.5 1.0 0.7 0.6 1.0 2.3
TRTWTRORGN 1.3 0.0 0.9 1.1 1.5 2.6 0.1 0.0 1.8 2.4 1.3
CNDTRTWTRORGN 0.1 0.5 1.8 1.4 0.6 1.1 0.2 2.9 2.6 0.8 0.7
*P (0.05e0.01), **P (0.009e0.001), ***P(<0.001).
Arcsin(y) transformation.
ln(yþ0.025) transformation.
O.W. Baughman et al. / Journal of Arid Environments 124 (2016) 193e204 197
no effect after March 2013 (Table 1). The pattern of reduced active
growth in die-off reversed in May 2013, when the die-off supported
signicantly more active seedlings than the control regardless of
other factors (Fig. 2). In the second growing season, there were no
signicant differences in number of active P. secunda between
control and die-off condition, regardless of other factors, but in May
2014 differences were nearly signicant (Table 1,P¼0.0552), with
more active P. secunda growing in the die-off (8.9% ±1.4% SE) than
control (3.3% ±0.6% SE, Fig. 2a).
Litter removal was associated with signicantly more P. secunda
active growth in May 2013 (raked: 15.2% ±1.9% SE, unraked:
5.2% ±1% SE) as well as throughout the second season, although
this was only true in the unwatered treatment in February and May
2014 (Table 1 TRTWTR interactions, not shown).
3.5. Effects of die-off and treatments on E. elymoides active growth
Similar to P. secunda, there were initially fewer E. elymoides
seedlings in the die-off, reversing to more active E. elymoides late in
the rst season as well as throughout the entire second season
(Table 2,Fig. 2b). The die-off supported signicantly fewer active
seedlings than the control when water was added only in
November and December 2012 and February and April 2013
(Table 2 CNDWTR interactions, Fig. 3). In November and
December, these effects were due to a strong germination response
of nonlocal seeds in the watered control and unwatered die-off,
respectively (Table 2 CNDWTRORGN interactions, data not
shown). There were no signicant between-condition differences
in the unwatered treatment in November 2012 and April 2013, and
signicantly more active seedlings in unwatered die-off than con-
trol in December 2012 and February 2013 (Table 2 CNDWTR in-
teractions). Water had no continuing effects after May 2013. Litter
removal had mixed effects on E. elymoides seedlings, with some
interactive effects with water addition and seed origin
(Appendix 2).
3.6. Effects of seed origin on P. secunda active growth
Cumulative emergence was signicantly higher for local
P. secunda seeds (Fig. 3a), and there were more actively growing
locally-collected seeds than nonlocal seeds for all but one of the
survey dates (November 2014, Table 1,Fig. 4), though this result
was only evident in the water addition treatments at some survey
Fig. 2. Proportion of viable seed showing active growth through two growing seasons
for P. secunda (a) and E. elymoides (b), showing differences between control (circles)
and die-off (triangles) as well as watered (black) and unwatered (white). Means and
standard errors are from untransformed data, and the inset gure magnies end-of-
season P. secunda results that are difcult to see at larger scale.
Fig. 3. Proportion of viable seeds that emerged in the rst growing season for
P. secunda (a) and E. elymoides (b) by experimental factors: control (C) and die-off (D)
condition, early-season watering treatment (W) and unwatered (X), raked (R) and
unraked (N) litter, and nonlocal (NL) and local (L) seed origin. Main effects are shown
when no interactions are present; * ¼P<0.05, and signicant interactions are indi-
cated by the 4-bar graph. Letters above bars indicate signicant differences based on
Tukey's HSD (P <0.05). Means and standard errors are untransformed data.
Fig. 4. Proportion of viable seed showing active growth through two growing seasons
from nonlocal (black) and local (white) seeds of P. secunda (diamonds) and E. elymoides
(squares). Means and standard errors are untransformed data, and the gure inset
magnies end-of-season results that are difcult to see at larger scale.
O.W. Baughman et al. / Journal of Arid Environments 124 (2016) 193e204198
periods, including early in the rst season (Nov 2012eFeb 2013)
and in May 2014 (Table 1 WTRORGN interactions, data not
shown). In May 2013, the difference in seedling activity between
raked and unraked treatments was three times larger for local than
nonlocal seeds, with higher activity in raked than unraked plots
(Table 1 TRTORGN, data not shown).
3.7. Effects of seed origin on E. elymoides active growth
Effects of seed origin on active growth were less consistent for
E. elymoides than for P. secunda, with more interactions with other
factors (Table 2). Cumulative emergence was signicantly higher
for nonlocal than local E. elymoides seeds (Table 2). Despite a series
of complex CNDWTRORGN interactions (data not shown) from
November 2012 through February 2013, the presence of active
seedlings in control and die-off did not differ dramatically by origin
until April and May 2013, when nonlocal plants exhibited signi-
cantly more active seedlings than local plants (Fig. 4). In November
2013 and February 2014, there were more nonlocal seedlings in the
control (nonlocal: 1.2e1.8% ±0.5% SE, local: 0.3e0.5 ±0.2% SE) but
fewer than local in the die-off (nonlocal: 5.9e7.6% ±1.3% SE, local:
9.7e11.3% ±1.8% SE; Table 2 CNDORGN interactions). Addition-
ally, in February 2014, there were more local than nonlocal plants in
unraked (nonlocal: 2.8% ±1% SE, local: 5.5% ±1.3% SE) but not raked
plots, and this pattern was also apparent in April 2014, but only in
die-off plots (unraked die-off nonlocal: 5.5% ±1.7% SE, unraked die-
off local: 13.1% ±2.4% SE; Table 2 TRTORGN and CNDTRTORGN
interactions, data not shown for three-way interaction). In May
2014, there were signicantly more local than nonlocal plants in
die-off but not control plots (die-off nonlocal: 5.6% ±1.2% SE, die-off
local: 9.8% ±1.8% SE; Table 2 CNDORGN interaction, Fig. 4) and in
unraked but not raked plots (unraked nonlocal: 2.6% ±0.8% SE,
unraked local 6.2% ±1.4% SE; TRTORGN interaction, data not
3.8. Effects of experimental factors on late-season vigor and growth
Seedlings of both P. secunda and E. elymoides in die-off plots
exhibited signicantly greater late-season vigor, more leaves, and
greater height than in control plots for both seasons, and more
individuals owered in die-off plots in the second season (Table 3,
Figs. 5 and 6). Local P. secunda exhibited signicantly greater late-
season vigor than nonlocal plants in both growing seasons,
though there were no differences in leaf number or plant height for
either season (Table 3,Fig. 5). Nonlocal E. elymoides plants exhibited
greater late-season vigor, more leaves, and greater height than local
in the rst season but not the second, where there were no sig-
nicant effects of origin on any of these responses (Table 3,Fig. 6).
3.9. Effects of B. tectorum dynamics on native emergence and
When B. tectorum density was included as a model covariate for
P. secunda, it showed a signicant positive relationship with
Table 3
Results of nonparametric Wilcoxonsigned rank tests for late-season vigor, number of leaves, seedling height, and number of plants with owering structures for P. secunda and
E. elymoides in both growing seasons. Values are
, and bolded effects are signicant (P <0.05). Due to small sample size, factors include only condition (CND, control vs. die-
off) and seed origin (ORGN, local vs. nonlocal), with no block effects or interactions included.
First season (2012e2013) Second season (2013e2014)
Late-season vigor No. of leaves Seedling height Late-season vigor No. of leaves Seedling height No. owering
P. secunda
CND 30.4*** 19.3*** 14.2*** 19.4*** 50.2*** 54.0*** 17.6***
ORGN 6.14* 0.17 0.78 16.8*** 0.78 0.83 0.19
E. elymoides
CND 43.8*** 53.9*** 77.8*** 7.30** 13.6*** 8.87** 20.2***
ORGN 7.96** 10.0** 13.5*** 1.65 1.66 1.23 2.92
*P (0.05e0.01), **P (0.009e0.001), ***P (<0.001).
Fig. 5. For P. secunda, late-season vigor (a, b) as measured by greenness index (0 ¼0%
green, 1 ¼1e25%, 2 ¼26e75%, 3 ¼76e100%), total number of leaves (c, d), plant
height as measured by longest leaf in centimeters (e, f), and proportion of viable seeds
that produced plants with owering structures (g) during the rst (left panels) and
second (right panels) growing seasons, by condition (control vs. die-off) and seed
origin (nonlocal vs. local). Asterisks indicate signicance based on Wilcoxon signed
rank tests (P <0.05). Values are plot means and standard errors.
O.W. Baughman et al. / Journal of Arid Environments 124 (2016) 193e204 199
emergence (F ¼6.6
,P¼0.0121) and April active seedling
number (F ¼7.3
,P¼0.0082), and a signicant negative cor-
relation with May active seedling number (F ¼4.3
The main effects of condition on April P. secunda active seedling
number (Table 1) were reduced when B. tectorum density was
included as a covariate (F ¼2.7
,P¼0.125), though the
CNDTRT interaction for April activity remained signicant
(F ¼4.2
,P¼0.043). No such correlations or results modica-
tions were observed for E. elymoides plots in response to B. tectorum
density as a covariate. When rst season B. tectorum biomass was
included in the model as a covariate, it showed a signicant positive
relationship with May active seedling number for E. elymoides
(F ¼5.9
,P¼0.0174), but main model results were not signif-
icantly modied.
4. Discussion
Restoration of highly-invaded arid systems can be extremely
challenging, especially when invaders exert strong effects on
community structure, disturbance regimes, and biogeochemical
processes (Brooks et al., 2004; Adair and Burke, 2010). Temporary
reductions in invasive species presence, through either manage-
ment actions or, in our case, the naturally-occurring phenomenon
of complete B. tectorum stand failure, or die-off, may improve the
restoration potential of these highly invaded systems if native
species can establish during these windows of opportunity. We
found lower cumulative emergence of two native perennial grasses
seeding into a die-off site than in the adjacent control, but seed-
lings that did emerge in the die-off exhibited signicantly greater
vigor, growth, and more individuals owered. Survival was also
improved in the die-off site: there were more actively growing
plants in the die-off at the end of both growing seasons for
P. secunda and throughout the whole second season for
E. elymoides. While previous work has examined the effect of die-
off on B. tectorum growth (Meyer et al., 2014; Nicholson, 2014),
this study is the rst to demonstrate the positive effects of die-off
on native species establishment. B. tectorum die-offs may represent
an opportunity for increasing the success of native seeds in highly
invaded systems.
We found higher N and P in die-off soil than in adjacent control,
which corroborates similar increases in N observed in previous
studies on B. tectorum die-offs (Blanketal.,2011;Meyeretal.,
2014), and removal experiments (Eckert and Evans, 1967; Eckert
et al., 1970). Additionally, we found die-off soil to be signicantly
wetter than adjacent control soil in winter and early spring,
especially when litter layers were left intact, and die-off soils were
also signicantly cooler than adjacent controls throughout the
rst growing season. These differences in soil conditions likely
played a role in the enhanced vigor, growth, and establishment
We also observed differences in growth and establishment
between locally collected and commercially produced, nonlocal
seeds, with differences varying by species and through time.
Local (P. secunda) had greater cumulative emergence, higher
numbers of active seedlings, and increased late-season vigor
through both growing seasons, consistent with trends of local
adaptation in many other plant species (Joshi et al., 2001; Leimu
and Fischer, 2008). In contrast, nonlocal E. elymoides (Toe Jam
Creek) seedlings had higher cumulative emergence, were more
abundant, vigorous, leaer, and taller than local seedlings in the
later part of the rst season, but this clear advantage of nonlocal
plants disappeared in the second season. In fact, by the end of the
second season, only local advantages were apparent, under spe-
cic conditions: there was greater survival of local E. elymoides
plants in die-off, but not control areas, and in unraked but not
raked plots. Increased seedling growth and vigor have been
highly prioritized traits in plant material development in the
region (Leger and Baughman, 2015), and this is true for Toe Jam
Creek (United States Department of Agriculture, 2013). In our
study, however, increased performance was not associated with
greater plant size, for either P. s ecu n da or E. elymoides. There is
evidence that increased rst year growth is independent of some
tness components, such as survival (Verdú and Traveset, 2005)
and fecundity (Chambers and Aarssen, 2009), and our ndings
substantiate other studies from the Great Basin which have
similarly shown that large above-ground size in herbaceous
plants is not necessarily related to increased survival (Kulpa and
Leger, 2013), especially in competitive environments (Rowe and
Leger, 2011).
We added early-season water and removed litter to see if these
treatments aided in the establishment of native seeds, in general, or
in sites that had experienced the die-off phenomenon. Although
cumulative emergence of both species was boosted by pre-
Fig. 6. For E. elymoides, late-season vigor (a, b) as measured by greenness index
(0 ¼0% green, 1 ¼1e25%, 2 ¼26e75%, 3 ¼76e100%), total number of leaves (c, d),
plant height as measured by longest leaf in centimeters (e, f), and proportion of viable
seeds planted that produced plants with owering structures (g) during the rst (left
panels) and second (right panels) growing seasons, by condition (control vs. die-off)
and seed origin (nonlocal vs. local). Asterisks indicate signicance based on Wil-
coxon signed rank test (P <0.05). Values are plot means and standard errors.
O.W. Baughman et al. / Journal of Arid Environments 124 (2016) 193e204200
germination water addition and unaffected by pre-planting litter
removal, these treatments had mixed effects on seedling perfor-
mance, but effects were mostly consistent between the die-off and
control. Water addition was generally unrelated to patterns of
P. secunda activity in either season, and litter removal was associ-
ated with higher P. secunda activity in the second but not the rst
season. For E. elymoides, water addition was associated with higher
activity in the rst but not the second season, and litter removal
had few but mixed effects throughout both growing seasons. In
short, native species did not show lasting benets of water addition
in the fall of the rst season, and removing litter appeared to
enhance survival of P. secunda. Given these ndings, litter reduction
may lead to improvements of some native species, in or out of a
recent die-off.
Competitive effects of B. tectorum on native Great Basin
perennial grasses can be substantial (Humphrey and Schupp,
2004; Melgoza et al., 1990), though some studies have found
such competition to be absent and/or much less important for
native performance than other factors (James and Svejcar, 2010;
Elseroad and Rudd, 2011; Mangla et al., 2011). In our study,
B. tectorum densities were higher in control, watered, and
unraked plots, while biomass was generally higher in the die-off,
and showed positive responses to water addition and litter
removal. Both measures of B. tectorum had positive, signicant
correlations with native active seedling growth in the rst
growing season, suggesting either facilitation or parallel re-
sponses to favorable plot conditions. Only one correlation, be-
tween B. tectorum density and P. secunda active seedling number
in May, was negative and suggestive of competitive effects, and
including B. tectorum dynamics in the model did not eliminate
the signicant effect of condition on native seed performance.
Thus, our correlations failed to suggest that differences in in-
teractions with B. tectorum were the main factors responsible for
increased native seedling performance in die-offs. Collective
B. tectorum density and biomass in both conditions during the
study were far below maximum possible values (Eckert and
Evans, 1967; Knapp, 1996), perhaps due to below-average pre-
cipitation or in response to die-off conditions, and this could
account in part for a lack of clear competitive effects.
Though B. tectorum stand failure was not observed in this study,
emergence and early-season active growth of native species was
generally lower in die-off than control plots for both species. This
pattern was not clearly affected by early-season water addition or
litter removal treatments, and the observed differences in soil
(slightly lower pH, and higher P and N in die-off than control)
would not be expected to directly affect emergence in this way. The
cooler and wetter conditions in die-off soils may have somehow
reduced emergence, either directly or through their effects on soil
pathogen activity, though this is only speculation. This study did
not directly investigate the causes of native seed mortality in the
former die-off area, and this should be a topic of future research,
especially given the otherwise positive effects of recent die-off sites
for the studied native species.
Finally, we note that the hand-sorting, precision-seeding, and
intensive monitoring methods employed in this study are not those
employed in large-scale restorations, and thus our establishment
rates are not directly comparable to studies reporting survival or
success of native perennial grasses in wildland settings. Seeding
methods play a large role in determining the patterns of success
when restoring resource-limited, semiarid sites (James and Svejcar,
2010; Bernstein et al., 2014) by determining the conditions that
germinating seeds will experience, which are perhaps the
conditions of most concern for improving restoration success in
this region (Boyd and Davies, 2012). Our future efforts will identify
the effects of recent B. tectorum die-offs on native establishment
using realistic, larger-scale restoration seeding methods.
5. Conclusions
In the face of one of the more signicant biological invasions of
the western United States, there is much to be gained by attempting
to reduce and reverse ecological degradation through native plant
restoration. Results from this northern Nevada die-off site indicated
some initial negative but overall positive effects of seeding in a die-
off area on the growth and establishment of two native grasses.
Additionally, this case study demonstrated source-related and
species-dependent differences in performance of native perennial
grasses seeded into an invaded site, highlighting the importance of
seed selection in restoration for the region. Our ndings demon-
strate that further investigations of the restoration potential of die-
offs in other areas are warranted, as results may differ as environ-
mental conditions vary between sites, or between years at the same
site. To maximize our understanding of the effects of die-offs on
native seed establishment, we recommend that future efforts
examine multiple die-off sites, manipulate B. tectorum densities
and soil resources, and test whether establishment benets are
observed on larger scales and when traditional restoration methods
are used.
This work was funded by the Bureau of Land Management
under the Integrated Cheatgrass Die-off Project, as well as by the
Nevada Department of Wildlife and the Great Basin Landscape
Conservation Cooperative (2013). We thank Drs. Lee Turner and
Matthew Forister for aiding in coordinating this project (LT) and
rening the manuscript (LT, MF). We thank Rob Burton and Dr.
Calvin Jennings of the Winnemucca District of the Bureau of Land
Management for enabling the establishment of the study site, and
L&H Seeds, Inc. for donating commercial seeds. Many skilled eld
assistants and volunteers made this project possible, including
Lyndsey Boyer, Bryce Wehan, Meghan Whitman, Molly Bechtel,
Brian McMillan, Jacob Brannam, Wailea Johnston, Brittany Trim-
ble, Riley Anderson, Brianna Kooreman, Curt Baughman, Dashiell
Hibbard, Michelle Hochrein, and Drs. Kevin Badik, Lauren Por-
ensky, and Dan Atwater.
Table A.1
Means ±standard errors of untransformed data, and ANOVA results for soil nutri-
ents (ppm) and organic matter (%) analysis in control and die-off soils. Asterisks
indicate signicant differences (P <0.05), and greater values are bolded when there
are signicant between-condition differences.
Fall 2012 Summer 2013
Control Die-off F
Control Die-off F
pH 6.9 ±0.05 6.52 ±0.05 42.4** 7.45 ±0.1 6.84 ±0.12 26.8***
P72.4 ±3.78 91.8 ±4.55 47.9** 68 ±7.62 116 ±5.55 25.2***
K 473 ±16 536 ±48.1 1.09 627 ±43.1 778 ±38.9 7.5*
Ca 1242 ±28.2 1119 ±88.9 1.64 1426 ±38.8 1271 ±48.5 4.01
Mg 243 ±6.22 210 ±18.4 2.66 330 ±12 275 ±10.4 7.3*
OM 2.52 ±0.15 3.08 ±0.12 5.44 1.66 ±0.06 1.7 ±0.13 0.06
7.8 ±0.49 18.8 ±1.24 80.7*** 3 ±0.39 11 ±2.98 8.8*
*P (0.05e0.01), **P (0.009e0.001), ***P (<0.001).
O.W. Baughman et al. / Journal of Arid Environments 124 (2016) 193e204 201
Appendix 1. Effects of die-off, litter removal and watering on soil
moisture and temperature
Litter removal and watering altered the effects of the die-off
on moisture in the winter, early spring, and late spring
(CNDWTR, TRTWTR and CNDTRTWTR interactions,
Table A.2). Soil moisture was slightly higher (0.002 cm
soil ±0.001) in winter and substantially higher (0.02 cm
soil ±0.001) in early spring in die-off than control soils
when unraked. Litter removal created relatively dry soils through
the season, especially in early spring following spring-thaw when
soils dried down more quickly. For example, by March, soil
moisture was at least 32% lower in control than die-off treatments
when raked and watered. Although watering reduced the differ-
ences between control and die-off conditions, from early spring
on, the driest soils were watered control and die-off plots that
were raked. Soil moisture in these two treatments was at least
25% lower than any other treatment combination. In the winter,
all treatment soils experienced a series of freezeethaw cycles that
caused soil moisture to freeze leading to a decrease in available
soil water. In early and late spring, soils experienced six dry-
ingerewetting cycles due to rainfall events beginning in early
Soil temperatures were inuenced by condition, litter removal,
and water addition during the entire rst season (Table A.1). Litter
removal and water altered the effects of the die-off on temperature
in early spring and late spring (Table A.2 CNDTRTWTR in-
teractions). In winter, temperatures were 0.78
C±0.06 (n¼21)
cooler in die-off than control soil but only during freezeethaw
cycles with the absence of litter further depressing soil tempera-
tures. Also, in early spring and spring, temperatures in die-off soil
were 1.93
C±0.09 (n¼90) cooler, and these differences were
accentuated when litter was intact. Generally, winter pulses of
water slightly decreased soil temperatures over the season
regardless of other factors.
Appendix 2. Effects of litter removal and water addition treatments
on E. elymoides active growth
In November 2012, signicantly fewer active E. elymoides
seedlings existed in raked than unraked treatment in the
control that was watered, while the unwatered control and
die-off of either watering treatment did not exhibit effects
of raking (Table 2 CNDTRTWTR interaction).
Fewer seedlings existed in the raked than unraked treatment
in December 2012 regardless of other factors (unraked:
34.5% ±3.8% SE, raked: 29% ±3.7% SE), as well as in February
2013 in the unwatered but not watered treatment (TRTWTR
interaction, data not shown). By May 2013, this pattern had
reversed, with the raked treatment showing higher activity
(unraked: 32.1% ±2.2% SE, raked: 50.4% ±2.3% SE). There were no
effects of litter treatment on number of active seedlings in March
or April 2013.
Water addition increased E. elymoides seedling activity
regardless of other factors from November 2012 through April
2013, with the effect larger in unraked than raked treatments
early in the season (Table 2 TRTWTR interactions, data not
shown). In May 2013, the watered treatment supported more
active seedlings than unwatered only when raked (watered:
61.8% ±2.4% SE, unwatered: 38.9% ±2.9% SE), with no differ-
ence when unraked (Tab le 2 TRTWTR interaction, data not
Table A.2
Repeated measures ANOVA results for soil moisture and temperature during the winter (1st Nov e28th February), early spring (1st March e30th April), and late spring (1st
May e31st May) in the rst season. Values are F statistics, df (treatment, error, presented in parentheses for moisture and arethe same for temperature), and bolded effects are
signicant (P <0.05). Factors include condition (CND, control vs. die-off), seedbed treatment (TRT, raked vs. unraked), and water treatment (WTR, watered vs. unwatered), and
Soil Moisture Soil temperature
Winter Early spring Late spring Winter Early spring Late spring
CND 240***
7.74** 2236*** 1792***
TRT 190***
3.70 345*** 40.7***
WTR 47.4***
96.6*** 30.7*** 70.7***
Time 64.5***
1870*** 127*** 392***
103*** 63.6*** 12.5***
CNDWTR 34.0***
2.93 9.97** 32.4***
CNDTime 0.99
23.8*** 6.69*** 5.84***
TRTWTR 63.1***
19.5*** 52.8*** 72.5***
TRTTime 1.34*
5.42*** 5.87*** 7.63***
WTRTime 1.02
0.33 0.32 0.76
5.00* 4.12* 4.89*
CNDTRTTime 0.20
0.58 0.38 0.44
CNDWTRTime 0.67
0.26 0.18 0.87
TRTWTRTime 0.19
0.22 1.05 0.32
0.14 0.19 0.11
*P (0.05e0.01), **P (0.009e0.001), ***P (<0.001).
O.W. Baughman et al. / Journal of Arid Environments 124 (2016) 193e204202
Adair, E.C., Burke, I.C., 2010. Plant phenology and life span inuence soil pool dy-
namics: Bromus tectorum invasion of perennial C(3)-C(4) grass communities.
Plant Soil 335, 255e269.
Baughman, O.W., Meyer, S.E., 2013. Is Pyrenophora semeniperda the cause of downy
brome (Bromus tectorum) die-offs? Invasive Plant Sci. Manag. 6, 105e111. http://
Baughman, O.W., 2014. Will Native Plant Succeed where Exotic Invaders Fail?
Cheatgrass Die-off as an Opportunity for Restoration in the Great Basin, USA
(M.S. thesis). University of Nevada, Reno.
Beckstead, J., Meyer, S.E., Connolly, B.M., Huck, M.B., Street, L.E., 2010. Cheatgrass
facilitates spillover of a seed bank pathogen onto native grass species. J. Ecol. 98,
Bernstein, E.J., Albano, C.M., Sisk, T.D., Crews, T.E., Rosenstock, S., 2014. Establishing
cool-season grasses on a degraded arid rangeland of the colorado plateau.
Restor. Ecol. 22, 57e64.
Bithell, S.L., Butler, R.C., Harrow, S., McKay, A., Cromey, M.G., 2011. Susceptibility to
take-all of cereal and grass species, and their effects on pathogen inoculum.
Ann. Appl. Biol. 159, 252e266.
Blank, R., Morgan, T., Clements, D., 2011. Cheatgrass dead zones in northern Nevada.
Proc. Soc. Range Manag. 64, 94.
Booth, M.S., Caldwell, M.M., Stark, J.M., 2003. Overlapping resource use in three
Great Basin species: implications for community invasibility and vegetation
dynamics. J. Ecol. 91, 36e48.
Boyd, C.S., Davies, K.W., 2012. Differential seedling performance and environmental
correlates in shrub canopy vs. interspace microsites. J. Arid Environ. 87, 50e57.
Brooks, M.L., D'Antonio, C.M., Richardson, D.M., Grace, J.B., Keeley, J.E.,
DiTomaso, J.M., Hobbs, R.J., Pellant, M., Pyke, D., 2004. Effects of invasive alien
plants on re regimes. Bioscience 54, 677e688.
Chambers, J., Aarssen, L., 2009. Offspring for the next generation: most are pro-
duced by small plants within herbaceous populations. Evol. Ecol. 23, 737e751.
Fig. A.1. The effects of the die-off condition and litter removal by early season watering treatment on soil moisture (a &b) and temperature (c &d). Values (n¼6) are real-time
sensor data expressed on a daily time step from a soil depth of 2 cm. The gap in the data from January 28th to February 10th was due to an electrical storm resetting the data loggers.
Soil in the die-off was continually wetter in winter (die-off: 0.095 cm
H2O cm-3 soil ±0.002 SE, control: 0.083 ±0.002 SE) and early spring (die-off: 0.081 ±0.002 SE, control:
0.066 ±0.002 SE), as well as cooler in winter (die-off: 2.29 C±0.17 SE, control: 2.35 ±0.19 SE), early spring (die-off: 10.00 ±0.22 SE, control: 11.43 ±0.25 SE), and late spring (die-
off: 18.06 ±0.21 SE, control: 20.28 ±0.25 SE).
O.W. Baughman et al. / Journal of Arid Environments 124 (2016) 193e204 203
Clausen, J., Keck, D.D., Hiesey, W.M., 1947. Heredity of geographically and ecologi-
cally isolated races. Am. Nat 81, 114e133.
Davies, K.W., Boyd, C.S., Beck, J.L., Bates, J.D., Svejcar, T.J., Gregg, M.A., 2011. Saving
the sagebrush sea: an ecosystem conservation plan for big sagebrush plant
communities. Biol. Conserv.144, 2573e2584.
Eckert, R.E., Evans, R.A., 1967. A chemical-fallow technique for control of downy
brome and establishment of perennial grasses on rangeland. J. Range Manag.
20, 35e41.
Eckert, R.E., Klomp, G.J., Young, J.A., Evans, R.A ., 1970. Nitrate-nitrogen status of
fallowed rangeland soils. J. Range Manag. 23, 445e447.
Elseroad, A.C., Rudd, N.T., 2011. Can imazapic increase native species abundance in
cheatgrass (Bromus tectorum) invaded native plant communities? Rangel. Ecol.
Manag. 64, 641e648.
Goergen, E.M., Leger, E.A., Espeland, E.K., 2011. Native perennial grasses show
evolutionary pesponse to Bromus tectorum (cheatgrass) invasion. PLoS One 6,
Gornish, E.S., Aanderud, Z.T., Sheley, R.L., Rinella, M.J., Svejcar, T., Englund, S.D.,
James, J.J., 2014. Altered snowfall and soil disturbance inuence the early life
stage transition and recruitment of a native and invasive grass in a cold desert.
Oecologia 177, 595e606.
Hardegree, S.P., Jones, T.A., Roundy, B.A., Shaw, N.L., Monaco, T.A., Briske, D.D., 2011.
Assessment of range planting as a conservation practice. In: Briske, D.D. (Ed.),
Conservation Benets of Rangeland Practices: Assessment, Recommendations,
and Knowledge Gaps. USDA-NRCS, Washington, DC, USA, pp. 171e212.
Hartman, G.L., Chang, H.X., Leandro, L.F., 2015. Research advances and management
of soybean sudden death syndrome. Crop Prot. 73, 60e66.
Humphrey, L.D., Schupp, E.W., 2004. Competition as a barrier to establishment of a
native perennial grass (Elymus elymoides) in alien annual grass (Bromus tecto-
rum) communities. J. Arid Environ. 58, 405e422.
James, J.J., Svejcar, T., 2010. Limitations to postre seedling establishment: the role
of seeding technology, water availability, and invasive plant abundance. Rangel.
Ecol. Manag. 63, 491e495.
Jensen, S., Stettler, J., 2012. In: Applying provisional seed zones to plant materials
development in the Great Basin, and cultural practices. Great Basin Native Plant
Selection and Increase Project Annual Meeting. UT, Salt Lake City.
Johnson, R.C., Hellier, B.C., Vance-Borland, K.W., 2013. Genecology and seed zones
for tapertip onion in the US Great Basin. Botany 91, 686e694.
Jones, T.A., Nielson, D.C., Larson, S.R., Johnson, D.A., Monaco, T.A., Caicco, S.L.,
Ogle, D.G., Young, S.A., Carlson, J.R., 2004. Registration of Toe Jam Creek
bottlebrush squirreltail germplasm. Crop Sci. 44, 1880e1881.
Joshi, J., Schmid, B., Caldeira, M.C., Dimitrakopoulos, P.G., Good, J., Harris, R.,
Hector, A., Huss-Danell, K., Jumpponen, A., Minns, A., Mulder, C.P.H., Pereira, J.S.,
Prinz, A., Scherer-Lorenzen, M., Siamantziouras, A.S.D., Terry, A.C.,
Troumbis, A.Y., Lawton, J.H., 2001. Local adaptation enhances performance of
common plant species. Ecol. Lett. 4, 536e544.
Kawecki, T.J., Ebert, D., 2005. Conceptual issues in local adaptation. Ecol. Lett. 7,
Knapp, P.A., 1996. Cheatgrass (Bromus tectorum L) dominance in the Great Basin
Desert eHistory, persistence, and inuences to human activities. Glob. Environ.
Change 6, 37e52.
Kulpa, S.M., Leger, E.A., 2013. Strong natural selection during plant restoration fa-
vors an unexpected suite of plant traits. Evol. Appl. 6, 510e523. http://
Lambert, S.M., Monsen, S.B., Shaw, N., 2011. Notice of Release of Mountain Home
Germplasm Sandberg Bluegrass (Selected Germplasm, Natural Track). U.S.
Department of Agriculture, Forest Service, Fort Collins, CO, p. 7.
Leger, E.A., Baughman, O.W., 2015. What seeds to plant in the Great Basin?
Comparing traits prioritized in native plant cultivars and releases with those
that promote survival in the eld. Nat. Areas J. 35, 54e68.
Leimu, R., Fischer, M., 2008. A meta-analysis of local adaptation in plants. PLoS One
3, 1e8.
Linhart, Y.B., Grant, M.C., 1996. Evolutionary signicance of local genetic differen-
tiation in plants. Ann. Rev. Ecol. Syst 27, 237e277.
Loveless, M.D., Hamrick, J.L., 1984. Ecological determinants of genetic structure in
plant populations. Ann. Rev. Ecol. Syst 15, 65e95.
Mack, R.N., 1981. Invasion of Bromus tectorum L into western North America ean
ecological chronicle. Agro-Ecosystems 7, 145e165.
Mangla, S., Sheley, R.L., James, J.J., Radosevich, S.R., 2011. Role of competition in
restoring resource poor arid systems dominated by invasive grasses. J. Arid
Environ. 75, 487e493.
Melgoza, G., Nowak, R.S., Tausch, R.J., 1990. Soil-water exploitation after re e
competition between Bromus tectorum (cheatgrass) and 2 native species.
Oecologia 83, 7e13.
Meyer, S.E., Beckstead, J., Allen, P.S., Pullman, H., 1995. Germination ecophysiology of
Leymus cinereus (Poaceae). Int. J. Plant Sci. 156, 206e215.
Meyer, S.E., Franke, J., Baughman, O.W., Beckstead, J., Geary, B., 2014. Does Fusarium-
caused seed mortality contribute to Bromus tectorum stand failure in the Great
Basin? Weed Res. 54, 511e519.
Nasri, M., Doescher, P.S., 1995. Effect of competition by cheatgrass on shoot growth
of Idaho fescue. J. Range Manag. 48, 402e405.
Natural Resources Conservation Service Soil Survey Staff, 2014. United States
Department of Agriculture. Web Soil Survey (accessed 04.01.14). Available on-
line at:
Nicholson, J.A., 2014. Cheatgrass Die-off Phenomena: What are the Short and Long
Term Recovery Factors of Bromus tectorum Stand Failure? (M.S. thesis) Brigham
Young University, Provo.
PRISM Climate Group, 2014. Oregon State University (accessed 12.02.14). Available
online at:
sek, P., Richardson, D.M., 2010. Invasive species, environmental change and
management, and health. Annu. Rev. Environ. Resour. 35, 25e55. http://
Rafferty, D.L., Young, J.A., 2002. Cheatgrass competition and establishment of desert
needlegrass seedlings. J. Range Manag. 55, 70e72.
Rice, K.J., Davis, C.A., 2009. In: Scales of Adaptation and Determining Seed Zones for
Native Grass Restoration in the Sierra Nevada. Oral Presentation. Ecological
Society of America Annual Meeting. Albuquerque, NM.
Rowe, C.L.J., Leger, E.A., 2011. Competitive seedlings and inherited traits: a test of
rapid evolution of Elymus multisetus (big squirreltail) in response to cheatgrass
invasion. Evol. Appl. 4, 485e498.
Rowe, C.L.J., Leger, E.A., 2012. Seed source affects establishment of Elymus multisetus
in postre revegetation in the Great Basin. West. North Am. Nat. 72, 543e553.
R Development Core Team, 2013. R: a Language and Environment for Statistical
Computing. R Foundation for Statistical Computing. Austria, Vienna. Available
online at.
Shepherd, D.N., Martin, D.P., vander Walt, E., Dent, K., Varsani, A., Rybicki, E.P., 2010.
Maize streak virus: an old and complex emergingpathogen. Mol. Plant Pathol.
11, 1e12.
Stevens, A.R., Anderson, V.J., Fugal, R., 2014. Competition of squirreltail with
cheatgrass at three nitrogen levels. Am. J. Plant Sci. 5, 990e996. http://
United States Department of Agriculture, 2013. Agricultural Research Service, Forest
and Range Research Laboratory. In: Plants for the West ePlant Materials
Release Catalog. Available online at:
Verdú, M., Traveset, A., 2005. Early emergence enhances plant tness: a phyloge-
netically controlled meta-analysis. Ecology 86, 1385e1394.
Wilcove, D.S., Rothstein, D., Dubow, J., Phillips, A., Losos, E., 1998. Quantifying
threats to imperiled species in the United States. Bioscience 48, 607e615.
O.W. Baughman et al. / Journal of Arid Environments 124 (2016) 193e204204
... in recent die-off soils than soils of nearby areas that did not experience die-off. [8][9][10] We conducted a precision seeding study to test how die-offs affected perennial grass seeds at a die-off in Pershing County, Nevada, in 2012. 5,8 We planted Sandberg bluegrass (Poa secunda) and bottlebrush squirreltail (Elymus elymoides) into a recent die-off as well as an adjacent intact cheatgrass stand (control). ...
... [8][9][10] We conducted a precision seeding study to test how die-offs affected perennial grass seeds at a die-off in Pershing County, Nevada, in 2012. 5,8 We planted Sandberg bluegrass (Poa secunda) and bottlebrush squirreltail (Elymus elymoides) into a recent die-off as well as an adjacent intact cheatgrass stand (control). The die-off supported more bluegrass and squirreltail through 2 years of monitoring, and seedlings in the die-off had significantly greater growth and vigor late in the growing season than those in the control. ...
... These findings show that the cheatgrass present in die-offs the year after the event is at a lower density but has higher individual vigor than in adjacent areas that did not die off and that these differences persist or fade away by the second year, depending upon the site, all of which corroborates previous findings. [8][9][10] Because of this quick recovery, the window for restoration benefits after die-off may be short. ...
Full-text available
• The phenomenon of cheatgrass die-off is a common and naturally occurring stand failure that can eliminate the presence of this annual grass for a year or more, affecting tens to hundreds of thousands of acres in some years. • We designed a study to determine if the temporary lack of cheatgrass caused by die-offs is a restoration opportunity. We seeded native perennial species at three die-offs in the Winnemucca, Nevada, area. • Native grass establishment in die-offs was almost three times higher in the first season at all sites, relative to adjacent areas without die-off. Establishment was five times higher in the die-off at two sites in the second season, and plants produced dramatically more culms in the die-off at the third site in the third season. • Increasing seed rates led to more seedlings establishing in both die-offs and controls, with the strongest effect in the second season. • We suggest that landowners and managers consider targeting die-offs as efficient locations to focus native restoration efforts and that restoration practitioners should consider increasing seeding rates to maximize success.
... A commonly observed but poorly studied phenomenon in cheatgrass monocultures is the intermittent occurrence of 'die-off' or stand failure, where a cheatgrass stand fails to re-establish or is destroyed prior to seed production, despite receiving adequate precipitation for establishment (Baughman et al., 2016;Blank et al., 2011;Meyer et al., 2014). Cheatgrass die-offs occur at multiple spatial scales and in contrasting ecological settings, suggesting that there may be multiple causal agents, both abiotic and biotic, and that these agents may interact in complex ways to produce the die-off effect. ...
... Die-offs sometimes occur over very large areas and can have negative consequences, including increased soil erosion and desertification, loss of forage for livestock and wildlife, and invasion by secondary perennial weeds. However, cheatgrass dieoff can also provide a restoration opportunity (Baughman et al., 2016). This might be especially so in die-offs that occur in the same area for multiple years, as the prolonged reduction in competition could result in significantly increased native plant performance (Baughman and Leger, pers. ...
... Cheatgrass die-off has been observed to occur more frequently in certain areas than in others (Baughman et al., 2016), but occurrence is highly stochastic and often in remote areas. On-the-ground detection of cheatgrass die-off is therefore unreliable and inconsistent, and remote sensing indicators are sorely needed to quantify die-off incidence over long time periods and across extensive regions. ...
The exotic annual grass Bromus tectorum (cheatgrass) dominates vast acreages of rangeland in the western USA, leading to increased fire frequency and ecosystem degradation that is often irreversible. Episodic regeneration failure (“die-off”) has been observed in cheatgrass monocultures and can have negative ecosystem consequences, but can also provide an opportunity for restoration of native species and ecological function within the imperiled sagebrush steppe ecosystem. Proximate causes of cheatgrass die-off are uncertain, although several taxa of fungal soil pathogens have been implicated. Die-off occurrence is stochastic and can occur in remote areas. Thus, developing remote sensing indicators that are repeatable over long time periods, across extensive regions, and at relatively fine spatial resolution would be beneficial for accurately pinpointing events.
... We selected data from individual species of interest (Table 1) and summarized the abundance of (1) a combination of 16 EAGs, (2) cheatgrass, (3) medusahead, and (4) Sandberg bluegrass for model development and validation ( Figure 1). Sandberg bluegrass is a native perennial bunch grass found in much of the western United States and Canada and has phenological traits similar to exotic annual grasses that might lead to mapping confusion [36]. To permit upscaling of the species abundance data to the entire study area, a suite of predictor variables was selected and input into our machine-learning framework. ...
Full-text available
The invasion of exotic annual grass (EAG), e.g., cheatgrass (Bromus tectorum) and medusahead (Taeniatherum caput-medusae), into rangeland ecosystems of the western United States is a broad-scale problem that affects wildlife habitats, increases wildfire frequency, and adds to land management costs. However, identifying individual species of EAG abundance from remote sensing, particularly at early stages of invasion or growth, can be problematic because of overlapping controls and similar phenological characteristics among native and other exotic vegetation. Subsequently, refining and developing tools capable of quantifying the abundance and phenology of annual and perennial grass species would be beneficial to help inform conservation and management efforts at local to regional scales. Here, we deploy an enhanced version of the U.S. Geological Survey Rangeland Exotic Plant Monitoring System to develop timely and accurate maps of annual (2016–2020) and intra-annual (May 2021 and July 2021) abundances of exotic annual and perennial grass species throughout the rangelands of the western United States. This monitoring system leverages field observations and remote-sensing data with artificial intelligence/machine learning to rapidly produce annual and early season estimates of species abundances at a 30-m spatial resolution. We introduce a fully automated and multi-task deep-learning framework to simultaneously predict and generate weekly, near-seamless composites of Harmonized Landsat Sentinel-2 spectral data. These data, along with auxiliary datasets and time series metrics, are incorporated into an ensemble of independent XGBoost models. This study demonstrates that inclusion of the Normalized Difference Vegetation Index and Normalized Difference Wetness Index time-series data generated from our deep-learning framework enables near real-time and accurate mapping of EAG (Median Absolute Error (MdAE): 3.22, 2.72, and 0.02; and correlation coefficient (r): 0.82, 0.81, and 0.73; respectively for EAG, cheatgrass, and medusahead) and native perennial grass abundance (MdAE: 2.51, r:0.72 for Sandberg bluegrass (Poa secunda)). Our approach and the resulting data provide insights into rangeland grass dynamics, which will be useful for applications, such as fire and drought monitoring, habitat suitability mapping, as well as land-cover and land-change modelling. Spatially explicit, timely, and accurate species-specific abundance datasets provide invaluable information to land managers.
... Stand failures represent a natural form of cheatgrass control and can provide an opportunity for native plant restoration (Meyer et al., 2014). For example, when native grass seeds were planted in a stand failure area, native grasses were able to outcompete cheatgrass in the following years (Baughman et al., 2016). Since stand failures were first observed in the 1930s, several hypotheses for the occurrence of stand failures have been put forth, ranging from abiotic factors such as weather to some different fungal agents such as Microdochium nivale and Ustilago bullata (Klemmedson & Smith, 1964;Meyer et al., 2010;Piemeisel, 1938). ...
Full-text available
Cheatgrass (Bromus tectorum L.) is an invasive annual grass (Poaceae) that has colonized large portions of the Intermountain West. Cheatgrass stand failures have been observed throughout the invaded region, the cause of which may be related to the presence of several species of pathogenic fungi in the soil or surface litter. In this metabarcoding study, we compared the fungal communities between sites that have and have not experienced stand failure. Samples were taken from the soil and surface litter near Winnemucca, Nevada, and in Skull Valley, Utah. Our results show distinct fungal communities associated with stand failure based on both geography and sample type. In both the Winnemucca and Skull Valley surface litter, there was an elevated abundance of the endophyte Ramimonilia apicalis in samples that had experienced a stand failure. Winnemucca surface litter stand failure samples had an increased abundance of a potential pathogen in the genus Comoclathris. Skull Valley surface litter stand failure samples had an increased abundance of an undescribed new species in the Rustroemiaceae family which is responsible for the so‐called bleach blonde syndrome in cheatgrass, while the soils had an increased abundance of potential pathogens in the genera Olpidium and Monosporascus. Metabarcoding via targeted ITS sequencing was used to characterize fungal communities in cheatgrass soils affected by stand failure in multiple areas. Analysis of the data confirmed key differences in the overall community composition between native communities and those affected by stand failure. Several identified fungal pathogens warrant further investigation to determine whether they are causal agents of stand failure.
... Dalea ornata , it is surprising that while these species presumably undergo a number of freeze-thaw cycles under field conditions (Baughman et al., 2016), they did not respond to any of the tested temperature-based treatments. This is contrary to findings observed in other members of the Fabaceae, which responded to wet heat Long et al., 2012;, freezing + wet heat , or freeze-thaw . ...
Full-text available
Direct seeding is a fundamental means of restoring degraded plant communities but often achieves less than 5% plant recruitment. In drylands, plant establishment is most constrained during germination and emergence. To promote plant recovery in highly degraded environments, seed treatments can be used to relieve dormancy and enhance the performance of difficult-to- ' establish species. Focusing on two biodiverse bioregions (the North American Great Basin and the Western Australian Pilbara), I resolved the primary seed germination traits of key regional plant species and demonstrated the potential of combining dormancy alleviation treatments with seed enhancement technologies to improve dryland restoration efforts.
... The lack of a late period reduction in E. elymoides biomass is due to ontogenetic dieback of living tissue. Similar to many C 3 grasses, E. elymoides achieves maximum biomass in April or May and has almost entirely senesced by August (James et al. 2008;Baughman et al. 2016). Thus, sensitivity to a second drought must be assessed earlier in the season for C 3 grasses like E. elymoides compared to C 4 species like B. gracilis. ...
Full-text available
Climate change will alter global precipitation patterns, making it increasingly important that we understand how ecosystems will be impacted by more frequent and severe droughts. Yet most drought studies examine a single, within-season drought, and we know relatively little about the impacts of multiple droughts that occur within a single growing season. This distinction is important because many plant species are able to acclimate physiologically, such that the effects of multiple droughts on ecosystem function deviate significantly from the effects of cumulative, independent droughts. Unfortunately, we know relatively little about the ability of dominant species to acclimate to drought in drought-sensitive ecosystems like semi-arid grasslands. Here, we tested for physiological acclimation to multiple drought events in two dominant shortgrass steppe species: Bouteloua gracilis (C4) and Elymus elymoides (C3). Neither species exhibited physiological acclimation to drought; leaf water potential, stomatal conductance, and photosynthesis rates were all similarly affected by a single, late period drought and a second, late period drought. Biomass was lowest in plants exposed to two droughts, but this is likely due to the cumulative effects of both an early and late period drought. Our results suggest that late period droughts do exert weaker effects on biomass production of two dominant shortgrass species, but that the weaker effects are due to ontogenetic changes in plant physiology as opposed to physiological acclimation against multiple droughts. As a consequence, current ecosystem models that incorporate grass phenology and seasonal physiology should provide accurate predictions of primary production under future climates.
... Genetic diversity in seed can buffer against environmental variabil- ity, which greatly affects seed germination and emergence, and this should be further investigated. Previous field research found vari- ation among species in seed type performance and attributed it to factors such as soil water content, drought, nutrient loadings and competition (Jacobson et al., 1984;Bugg et al., 1997;Bischoff et al., 2010;Walker et al., 2014;Baughman et al., 2016). Any of the three hypotheses Wilsey (2010) proposes may be valid to explain cultivar and wild seed performance on a restoration site given that response is impacted by species or ecotype and site specific conditions. ...
Availability of native plant materials for grassland restoration is limited. Even when available in sufficient quantities for ecological restoration projects, seed germination and establishment in relatively arid environments is often low. Poor revegetation results in soil erosion, invasion by non-native plant species and reduced aesthetics. Therefore, development and use of native plants bred for traits favourable for restoration should be considered. This study addressed whether native cultivar seed, commercially selected for advantageous growth characteristics, could improve native grass species reestablishment relative to wild collected seed. Cultivar and wild seed types of four cool season native grass species were investigated at three foothills fescue grassland reclamation sites: Bromus carinatus (mountain brome), Elymus trachycaulus (slender wheatgrass), Festuca idahoensis (Idaho fescue) and Koeleria macrantha (June grass). Seeding and transplanting were conducted and germination, emergence, density, height and health were determined from 2011 to 2013. No significant differences were detected between cultivar and wild seed types except laboratory germination, which was greater in the Elymus trachycaulus cultivar and in the wild collected Koeleria macrantha. Bromus carinatus performed poorly as a seedling from either seed type. Consistent trends in cultivar and wild seed performance, that reflected seed germination, were found for each species although results were not significant due to high variability. Results show that for these common grass species, seed type may not influence initial establishment. Differences among species were significant and varied with response measured, suggesting species characteristics are a key factor affecting native grass reestablishment. Species specific responses to seed type highlight the importance of making seed source decisions on a species basis.
... Herget et al. [65] also identified differences in germination timing, early, and total root growth among our local collections, MT-1, High Plains and Opportunity. Along similar lines, Baughman et al. [66] compared local P. secunda collections with the Mountain Home release in the context of cheatgrass die-offs and found better performance in the local collections. Johnson et al. [63] found higher survival and greater leaf area in cultivars (of the variant 'Sandbergii') than in other collections in common gardens. ...
Full-text available
The genetics of native plants influence the success of ecological restoration, yet genetic variability of local seed collections and commercial seed releases remains unclear for most taxa. Poa secunda, a common native grass species in Intermountain West grasslands and a frequent component of restoration seed mixes, is one such species. Here, we evaluate the genetic variation of local Poa secunda collections in the context of wild populations and commercial seed releases. We evaluated AFLP markers for seven Poa secunda collections made over a 4000-hectare area and four commercial releases (High Plains, MT-1, Opportunity , and Sherman). We compare the genetic distance and distribution of genetic variation within and between local collections and commercial releases. The extent and patterns of genetic variation in our local collections indicate subtle site differences with most variation occurring within rather than between collections. Identical genetic matches were usually, but not always, found within 5 m 2 collection sites. Our results suggest that the genetic variation in two Poa secunda releases (High Plains and MT-1) is similar to our local collections. Our results affirm that guidelines for Poa secunda seed collection should follow recommendations for selfing species, by collecting from many sites over large individual sites.
Following removal of the invasive species Rhododendron ponticum, the native understorey plant community typically fails to reestablish itself. Potential explanations for this failure include (1) lack of an appropriate native seed source; (2) inability of seed to penetrate a dense bryophyte layer; and (3) persistence of chemical “legacy effects” in the soil. We established an experiment to test these competing hypotheses in an Atlantic oak woodland where R. ponticum had been removed. The following experimental treatments were applied singly and in combination: (1) addition of a native seed mix to test for seed limitation; (2) removal of the established ground vegetation at the start of the experiment (which principally consisted of bryophytes) to test for the impact of a barrier layer; (3) addition of activated carbon to test for chemical legacy effects in the soil; and (4) fertilization as an additional measure to promote the establishment of native vascular plants. Application of the native seed mix was revealed to be an effective way to increase the cover of native vascular plants and was particularly effective when applied after the removal of the bryophyte layer. The application of activated carbon and/or fertilizer, however, had no effect on the cover of native vegetation. We conclude that reports of R. ponticum exerting chemical legacy effects long after its removal may have been overstated and that seed limitation and inability to successfully establish in a dense bryophyte layer provided the strongest barriers to natural recolonization by the native plant community following R. ponticum removal.
Plants seeded during ecological restoration sometimes persist but more often fail to establish. Biodiversity has been shown to stabilize a number of ecological processes, suggesting biodiverse seed mixes could be designed to stabilize plant establishment outcomes. In particular, it may be possible to design seed mixes to increase chances at least some seeded species will be adapted to whatever environmental conditions arise during establishment. To explore this possibility, we developed a modelling framework and applied it to data from 30 field experiments (15 sites × 2 seeding years) conducted in a big sagebrush (Artemisia tridentata Nutt.) ecosystem. In each experiment, three native and one nonnative grass were sown (600 seeds m⁻²) in separate plots, and we estimated the probability each species germinated and survived through two growing seasons post-seeding. Applying an optimization algorithm to these survival probabilities allowed us to assign species identities to 600 seeds m⁻² in a manner maximizing the number of experimental conditions yielding ≥5 plants m⁻², a common plant density goal in grassland restoration. Allocating 353 (216, 555) [point estimate (95% CI)] seeds to Poa secunda J. Presl and 247 (11, 378) seeds to Pseudoroegneria spicata (Pursh) Á. Löve) maximized our native plant density goal (goal achieved in 12 (10, 14) of 30 experiments), and the allocation to >1 species supports the hypothesis biodiverse seed mixes could be designed to reduce establishment failures. Averaged over experiments, P. spicata survival was roughly half of P. secunda survival, but P. spicata nevertheless contributed to the density goal by compensating for low P. secunda survival in certain experiments. Strategically combining species with different seed/seedling traits can increase chances of achieving adequate plant establishment during ecological restoration.
Full-text available
Postfire revegetation with native perennial grasses is difficult to achieve in disturbed arid rangelands. If local populations are adapted to current conditions, then locally collected seed would be predicted to have higher survival than nonlocal seed, and using local seed should improve revegetation success. However, for revegetation projects in the Great Basin, sufficient quantity of local seed is often difficult to obtain commercially, so seeds often originate from source populations that are hundreds of kilometers from the project site. We investigated whether seed source affected first-year establishinent of big squirreltail (Elymus multisetus M.E. Jones) seedlings in a common garden field trial 50 km north of Reno, Nevada. For the trial, we used wild, locally collected seed and commercially produced seed originating from Oregon, Idaho, and California. Several phenological and growth traits varied significantly between source populations. Eighty-six percent of local seeds emerged, compared to 71%, 61%, and 12% of seeds from Idaho, Oregon, and California, respectively. Local seeds emerged, on average, 9 days earlier than seeds from other sources. Fourteen percent of the local seedlings survived through the first year, exceeding survival by Oregon (12%), Idaho (8%), and California (2%) seedlings. Though survivorship was highest for local seed, local seedlings were smaller, producing 24% fewer leaves than the most productive seedlings from the Idaho seed source. Our data suggest that seed source is an important factor in seedling establishment. If local seed can survive significantly better than regionally collected, commercially produced seed, it may be both ecologically and economically beneficial to use local seed in revegetation.
Full-text available
Restoration in the Great Basin is typically a large-scale enterprise, with aerial, drill, and broadcast seeding of perennial species common after wildfires. Arid conditions and invasive plants are significant barriers to overcome, but relatively simple changes to seeds used for restoration may improve success. Here we summarize: 1) the composition of seed mixes used in recent postfire seedings in Nevada, 2) traits that were valued when cultivars and other native seed materials were named and released, and 3) traits that have been demonstrated to increase native perennial grass performance in invaded systems. A review of 420 seeding treatments on public shrublands in Nevada between 2006 and 2009 indicated that native perennial grasses and native shrubs were most frequently included in these projects, followed by exotic and native forbs, and lastly, exotic perennial grasses. Native perennial grasses made up the bulk of seeds used in these treatments, with multiple species of grasses (average of 3.4 species) typically seeded per treatment, while the richness of other functional groups in seed mixes was closer to 1 species per treatment. Traits prioritized in cultivars and native seed material releases included, in order of frequency: forage quality and yield, seed yield, seedling vigor, ability to establish and persist, and drought tolerance, with many other traits mentioned with less frequency. Traits that had consistent support for improving native perennial grass performance in the field were related to early phenology, small size, and higher root allocation. Further tests to determine which traits improve shrub and forb establishment under field conditions could further refine seed source selection, and help maintain diversity in Great Basin systems.
Full-text available
Downy brome (cheatgrass) is a highly successful, exotic, winter annual invader in semi-arid western North America, forming near-monocultures across many landscapes. A frequent but poorly understood phenomenon in these heavily invaded areas is periodic 'die-off' or complete stand failure. The fungal pathogen Pyrenophora semeniperda is abundant in cheatgrass seed banks and causes high mortality. To determine whether this pathogen could be responsible for stand failure, we quantified late spring seed banks in die-off areas and adjacent cheatgrass stands at nine sites. Seed bank analysis showed that this pathogen was not a die-off causal agent at those sites. We determined that seed bank sampling and litter data could be used to estimate time since die-off. Seed bank patterns in our recent die-offs indicated that the die-off causal agent does not significantly impact seeds in the persistent seed bank.
Downy brome was controlled with three soil-active herbicides: atrazine, EPTC, and IPC. Seedings were made 1 year after herbicide application. If fallow were effective during this year, soil moisture was conserved. Seeding in deep furrows resulted in superior seedling stands and greater 2nd and 3rd year production than did surface drilling. Performance of Amur intermediate wheatgrass was superior to Standard crested and Topar pubescent wheatgrasses.
Ability to compete with alien weeds may be one factor enabling high-seral, native bunchgrasses to persist on degraded rangelands. This study examined the effect of competition from cheatgrass (Bromus tectorum L.) on shoot growth of Idaho fescue (Festuca idahoensis. Elmer). Four Idaho fescue collections were obtained from degraded rangelands, while the fifth was from a site in high ecological condition. Plants were established in pots in a greenhouse with 2 watering regimes, and ratios of Idaho fescue:cheatgrass of 1:0, 1:5, and 1:10. Plants were grown for 56 days. Increasing competition from cheatgrass depleted soil moisture and reduced growth of Idaho fescue. However, Idaho fescue produced greater tiller and leaf numbers than cheatgrass. Idaho fescue plants from the pristine population produced 0.57 g aboveground biomass while plants from the degraded sites produced 0.31 g. Aboveground biomass from the pristine population was reduced 35% and 56% at the 1:5 and 1:10 competition levels respectively, compared to the control (1:0 ratio). Aboveground biomass of plants from the degraded populations was similar to the control at the 1:5 level, and was reduced 32% at the 1:10 level. These results indicated that Idaho fescue from the degraded sites exhibits a different response to competition from cheatgrass than Idaho fescue from the pristine site. This information may prove useful in selecting ecotypes of Idaho fescue for range revegetation.
Nitrate-nitrogen $({\rm NO}_{3}\text{-}{\rm N})$ accumulated in the soil during the spring, summer, and fall of a fallow year. NO3- N levels in the surface 6 inches in fall, 1967 and 1968, were similar and averaged 43 lb./acre on the atrazine fallow, 27 lb./acre on the mechanical fallow, and 5 lb./acre on the check. Above average precipitation during the winter of 1968-69 resulted in less NO3- N in spring, 1969 compared to spring, 1968. A comparison between the 2 years at one location showed the following NO3- N levels in the surafce 6 inches: spring, 1968-atrazine fallow 30 lb./acre, mechanical fallow 29 lb./acre, and check 13 lb./acre; spring, 1969-atrazine fallow 5 lb./acre, and check 2 lb./acre.
Fusarium virguliforme causes soybean sudden death syndrome (SDS) in the United States. The disease was first observed in Arkansas in 1971, and since has been reported in most soybean-producing states, with a general movement from the southern to the northern states. In addition to F. virguliforme, three other species, Fusarium brasiliense, Fusarium crassistipitatum, and Fusarium tucumaniae, have been reported to cause SDS in South America. Yield losses caused by F. virguliforme range from slight to 100%. Severely infected plants often have increased flower and pod abortion, reduced seed size, increased defoliation, and prematurely senescence. Foliar symptoms observed in the field are most noticeable from mid to late reproductive growth stages. To manage SDS, research on crop rotations, soil types, tillage practices, seed treatments, and the development and utilization of host resistance has been investigated. This review focuses on what is known about F. virguliforme, the management of SDS in the United States, and how genetic engineering along with other traditional management options may be needed as integrated approaches to manage SDS.